Ruggedized Handheld
Computers in Biological
Research
Jeffrey Muday
David J. Anderson
Wake Forest University
Department of Biology
Why PDA / Handhelds?
• Automation of repetitive observational tasks
– Replacement of paper log notebooks?
– Increase accuracy of reporting tasks
– Reduce translational errors from traditional paper journals to
various electronic forms
– Save time
• Innovative research tools
– Handheld (PDA) computing form factor has many advantages
• Easier travel (customs)
• Small size / low weight
– Integration of digital data from various pieces of technology
(immersion monitors, PDA, GPS) centered on laptop as
analysis/reporting tool
– Integration of “interesting” peripherals
• Barcode scanning, GPS loggers, other data collection devices
Symbol 2700 in the Galapagos
Mark Westbrock
Jeff’s Goals
• Production of a “General Purpose” PDA
Database/Logging Program for use in classroom
and or research scenarios
• Partnership with ISSC R&D Group
expertise/resources to specific research domain
issues
• Development of domain specific tools as “Proof
of Concept”
• Maybe? Expand University’s Expertise in
innovative/novel technology and applications
PDA Use History
• PDA first suggested by D Anderson in 1997 as
a way of automating and standardizing logs
• WinCE platform suggested by a Computer
Science undergraduate, Chris Marts
• Investigation of PalmOS vs WinCE (1998)
• First PDA census program 1998-99
– Everex Freestyle-66 MHz processor and 8 MB of
RAM
• PDA proved difficult to interface to satellite
phones—needed a laptop in the field
Problems with Developing for PDA
•
•
•
•
•
•
Small memory
No true mass storage options
Lack of non-volatile storage
Battery life issues
Small screen size
Transfer of data from PDA to other platforms can
be challenging
• WinCE/PocketPC stripped of some very useful
classes (programmer’s problem)
Advantages of PDA
• Small size
– Convenient, portable
– Unobtrusive
– Can be worn on body
• Low power consumption
– (especially the Symbol 2700)
• High availability
• Durability
• Pen/Touchscreen interface
Census / Birdtrack
• Albatross Census; (1998-present)
–
–
–
–
–
–
–
Based on WinCE 2.0 VC++ platform
Everex Freestyle/Casiopeia E11 handheld
Observer log in
Band number typed on keypad
Lookup on limited database of existing birds
New data logged about a particular bird
Many iterations/variants have been used
• Foot damage, blood samples, sex confirmation, last seen
Screen Shots: The Census Program
Emphasis on Speed and Simplicity
Bird Carried GPS Logger
• GPS movement
downloader (2002)
– Downloads data from a
bird GPS “backpack”
– Can also be used to
download data from bird
“Immersion Monitors”
– Supports simple serial
(RS-232) standard
download with hardware
handshaking
Behavior Logging
• Nest Logger I; (2000-2002)
– Logging of time and actions by a particular
nest/family group
– Simple, fixed pushbutton interface
– Logging to plain-text date-based files
– The basis for a new “generic” Ethogram
program currently in writing stages for Wm.
Conner
Elaine Porter using B-Logger
B-Logger
• B-LOGGER I and II; (2002-2003)
– Based on Symbol Systems PDA platform (thanks to technical
and financial support from J. Dominick, A. Bishop, and P. Escott)
– Map-based Behavior logger (thanks C. Fulp for advice/support
on graphics)
– Screen gesture selects a location on a map accurate to 0.5m or
less, judged by accurate listings of hundreds of nest sites.
– Band numbers entered via keypad or laser barcode reader
(thanks to M. Johnson and R. McCartney)
– Excellent confidence going into the field because of battery
life/duty cycle research performed by MJ and RM.
– Subsequent release II in 2003 (for E Porter) supports larger
maps and better accuracy, variable screen origin, and zoomin/zoom-out capability.
Classroom EthoLog
Ethogram logger concept, Bill Conner
– Allows user to log behavioral events from a
list of choices
– Allows user to optionally identify the “actor(s)”
– Allows user to add new behaviors to the
predefined list of behavioral events
– Proof of concept established by Anderson’s
Nest Logger program
– Protoype was successfully piloted summer
session II by WFU students
A Sample Ethogram
Ethogram Button Wizard
Developed by Michael Swofford
Results
• Confidence:
– PDAs are rugged enough for field use
– battery life sufficient for frequent field activation of
Barcode scans
– Flexible enough for Domain specific application
• Time saved:
– Massive amounts of time saved by having logs in
electronic form
– Standardization of logging formats
– Frequent repetitive reliable observations now possible
• Improved Accuracy
Valuable Code Re-use
•
•
•
•
•
•
Keypad
Barcode
Map
Resource/INI file
Log file
Serial communication
Remaining Goals
• Integration of GPS
– To supply accurate Time, Date, Lat, Long, Elevation
coords.
– As context source for Map object
• Development of “Wizards” to customize
Ethogram program suite
– Valuable concept for any Database tool
• Development of a Biological/Archeological
collection tool (application: Botanical/Herbarium)
• Generic database/logger idea?
– Is a useful implementation possible or is the PDA too
“brittle” of a platform?
Future Directions
• Wireless networks
– Enhancement of communication
– Client/Server model
– “Uplink” of field data
• GPS awareness
– positional context sensitive database
– Ability to use PDA as a mapping device
• Tablet form-factor computing
– PDA is too small for some aspects of computing
– Integration with newer communication standards
Client/Server Computing
Problem: Databases will contain too much
data to be hosted on handheld computers.
Solution: Bring a database server to the field
and have PDAs function as clients.
Requirements
• Headless – requires no monitor keyboard
or mouse.
• Network – wireless 802.11x + Cat 5
Ethernet
• Removable solid-state mass storage
• Low power consumption
• Interconnectivity to PocketPC & Windows
Platforms (network + USB hotsync)
• Database hosting capabilities
Using an Old Sun Sparc Station
Fanless Motherboard for Field Station
Compact Flash as Mass Storage Device
Screen Captures of Embedded Computer